Theor. Appl. Climatol. (2008)
DOI 10.1007/s00704-007-0370-9
Printed in The Netherlands
1
WSL Swiss Federal Research Institute, Lausanne, Switzerland
Meteo-France, Illkirch-Graffenstaden, France
3
MeteoSwiss, Geneva, Switzerland
2
An analysis of the July 2006 heatwave extent in Europe
compared to the record year of 2003
M. Rebetez1 , O. Dupont2 , M. Giroud3
With 3 Figures
Received 27 April 2007; Accepted 3 November 2007; Published online 18 February 2008
# Springer-Verlag 2008
Summary
1. Introduction
Recent analyses have identified summer warming trends in
Europe in recent decades, culminating in 2003, when mean
summer temperatures were exceptionally hot over much of
Europe. Mean monthly temperatures were very high in July
2003 and reached record levels in both June and August.
In 2006, the mean monthly temperature for July reached
a record high. Our analysis of temperature observations
shows that in July 2006, as in summer 2003, maximum
temperatures were more abnormal than minimum values.
The 2006 heatwave was located more to the north than in
2003, and particularly affected the Netherlands, Belgium,
Germany, Poland, France and Switzerland. The July 2006
anomalies were similar in magnitude to those of June and
August 2003, but the discrepancy between minimum and
maximum temperature anomalies was larger in 2006
compared to both June and August 2003. For maximum
temperature, the affected land area by anomalies higher
than 4–6 K was largest in July 2006, although the anomalies
were higher in June and August 2003 at the most anomalous sites. In the north of Europe, the absolute monthly
temperature values were higher in July 2006 compared
to both June (also on the Iberian Peninsula) and August
2003.
More hot days and high-temperature events are
expected over nearly all land areas as a consequence of anthropogenic climate change (e.g.,
Easterling et al. 2000a, b, c; Barnett 2001;
IPCC 2001; Sparks et al. 2002; Bell et al.
2004; Rebetez 2004; Schaer et al. 2004; Stott
et al. 2004; IPCC 2007). Extreme events could
potentially have an enormous impact on the
economy, on human health, and on environmental sustainability. Abrupt changes in the
frequency, location or intensity of extreme meteorological events (floods, drought and storms,
etc.) have direct and severe effects on ecosystems and society, resulting in social and economic costs (e.g., Meehl et al. 2000; Parmesan
et al. 2000; Murnane 2004; Vescovi et al. 2005).
The need to plan for the effects of climate
change is becoming urgent for decision-makers
(Smoyer 1998; Kunkel et al. 1999; IPCC 2002;
Schnur 2002; Menne et al. 2003; De et al.
2004; Haines and Patz 2004; McMichael
and Woodruff 2004; McMichel et al. 2004;
Stephenson 2004; Kovats and Haines 2005;
IPCC 2007). Climate synopses and analyses
of recent events could provide us with valuable
Correspondence: Martine Rebetez, WSL Swiss Federal Research
Institute, CP 96, 1015 Lausanne, Switzerland, e-mail: rebetez@
wsl.ch
M. Rebetez et al.
knowledge about climate change directly relevant to decision-makers.
An unprecedented heatwave affected large
parts of the European continent in summer 2003
(Fink et al. 2004; Luterbacher et al. 2004;
Rebetez 2004; Schaer et al. 2004; Trigo et al.
2005; Rebetez et al. 2006; Della-Marta et al.
2007b). Summer 2003 was the hottest since at
least 1500 in Switzerland (Luterbacher et al.
2004) and at least 1761 in Germany (Sch€
onwiese
et al. 2004). More generally, recent analyses
show that over the period 1880–2005, the frequency of hot days has almost tripled over western Europe (Della-Marta et al. 2007a). Analyses
of the temperatures in summer 2003 have shown
that the record-breaking values were mainly
due to extremely hot maximum temperatures in
June and August (Rebetez et al. 2006). The heatwave was centred over France, Germany and
Switzerland. In July 2003, contrary to June and
August, temperatures were above normal, but
not record-breaking, even at the most anomalous
sites (Schaer et al. 2004; Rebetez et al. 2006).
As mean July temperatures were extremely high
in 2006, we analyzed maximum, minimum and
mean monthly anomalies over Europe. We compared our results with monthly data from June
and August 2003 in order to better understand
the heatwave of July 2006. We compared the
spatial extent of the anomalies above three temperature thresholds in order to show the differences between June 2003, August 2003 and July
2006 on one hand and the differences between
the anomalies for mean, minimum and maximum temperature on the other. We expect the
results to contribute to a better understanding
of these heatwaves in the context of climate
change.
2. Data and methods
We used mean, minimum and maximum monthly
surface air temperature data over Europe provided by Meteo-France and MeteoSwiss. We
elaborated anomaly data based on a 1961–1990
reference period. We mapped monthly mean,
minimum and maximum air temperature anomalies and compared the continental areas concerned by anomalies reaching three thresholds
(4=5=6 K) in June 2003, August 2003 and July
2006. We also mapped the absolute temperature
differences between July 2006 on one hand, and
June and August 2003 on the other.
Temperature data originated from CLIMAT
codes except for France and Switzerland.
CLIMAT codes are part of the basic data sets
and products supporting WMO programs and
which members exchange free of charge and free
of rights. French and Swiss data originated from
synoptic surface stations, so that extra data were
available compared to CLIMAT codes only. Quality control issues were applied systematically.
Data were selected from a domain 5 larger on
every side than the final mapped area in order to
keep a high spatial quality on the map edges. The
monthly data was systematically checked for
significance over the investigated area.
Temperature data were gridded for the whole
data selection zone using Point Kriging. A 25class isotropic experimental variogram was created, then a linear variogram model was chosen
and fitted to the experimental variogram curve.
The grid spacing value was set to 0.1 . Duplicate
data were defined by setting an X and Y tolerance to 0.02 . Only the point with the median Z
value was kept for each set of duplicates.
Original monthly data and mean monthly
values computed from minimum and maximum
temperature data were first mapped and analyzed
separately to detect visibly aberrant data. This
led us to eliminate a few rare meteorological stations showing doubtful measurement quality. The
final maps were established by again kriging
monthly and mean monthly temperatures separately, this time only considering the stations accepted for the analysis.
3. Results
Compared to 2003, the heatwave of July 2006
was centred more to the north, over Germany,
Belgium and the Netherlands (Fig. 1). The most
extreme anomalies were lower in July 2006 compared to June (minimum and maximum temperature) and August (maximum temperature)
2003. Maximum temperature anomaly values
above 8 K were measured in France in August
2003 and in France, Switzerland, Southern
Germany and central Italy in June 2003. Mean
temperature anomaly values above 6 K were
measured in France and Switzerland in June
2003 and in Germany in July 2006. Minimum
An analysis of the July 2006 heatwave extent in Europe compared to the record year of 2003
Fig. 1. Surface air temperature anomalies in July 2006 (top), June 2003 (middle) and August 2003 (bottom): mean (left),
minimum (middle) and maximum (right) surface air temperature
temperature anomaly values above 6 K were
measured in France, Switzerland and Italy in
June 2003.
Comparing the extent of the areas with
anomalies above 3 different thresholds reveals
that in July 2006, the mean monthly maximum
temperatures were more than 6 K above longterm average values over 686,000 km2 and an
area of 2,210,000 km2 was affected by maximum temperature anomalies higher than 4 K
(Table 1). For the same temperature thresholds,
the anomalies concerned smaller areas for
minimum and mean temperature. The areas affected by the three threshold anomalies for maximum temperatures were largest in July 2006
compared to June or August 2003 (Table 1 and
Fig. 1). For minimum temperatures, the areas
affected were largest in June 2003 and sec-
Table 1. European continental areas [km2] where the specified temperature anomaly thresholds (compared to the 1961–
1990 CLINO values) are exceeded in selected summer months
Maximum
temperature
>4 K
>5 K
>6 K
July 06
June 03
August 03
Minimum
temperature
2.21 106
1.95 106
1.77 106
>4 K
1.37 106
1.28 106
1.20 106
>5 K
6.86 105
6.41 105
5.89 105
>6 K
July 06
June 03
August 03
Mean temperature
4.66 105
1.14 106
5.52 105
>4 K
9.27 103
4.04 105
6.18 104
>5 K
0.00
4.67 104
0.00
>6 K
July 06
June 03
August 03
1.36 106
1.63 106
1.37 106
6.45 105
8.95 105
5.29 105
5.38 103
2.32 105
1.46 105
M. Rebetez et al.
ond largest in August 2003. For mean temperatures, the areas were largest in June 2003,
second largest in July 2006 for the 5 K threshold and in August 2003 for the 6 K and 4 K
thresholds.
Comparing minimum and maximum temperatures for each month shows that the greatest
anomalies were always reached for maximum
temperatures, with lesser anomalies occurring
for minimum temperature values (Fig. 1 and
Table 1). The discrepancy between minimum
and maximum temperature anomalies was larger
in 2006 compared to both June and August 2003
(Table 1).
Considering the absolute air temperature
values, July 2006 was hotter than June 2003
north of 45 N, as well as on the Iberian
Peninsula and Western France (Fig. 2). This fea-
Fig. 2. Differences between July 2006 and June 2003 (top)=August 2003 (bottom) average (left), minimum (middle) and
maximum (right) surface air temperature
Fig. 3. Statistical distribution of minimum (left) and maximum (right)
July monthly surface air temperatures
1895–2006 in France (60–68 stations).
Only stations below 500 m asl were
used
An analysis of the July 2006 heatwave extent in Europe compared to the record year of 2003
ture was more pronounced for maximum than for
minimum temperature. July 2006 was also hotter
than August 2003 over North-European lands
north of 50 N. Differences were stronger for
June than for August, and for maximum than
for minimum temperature. The 2006 heatwave
was clearly located more to the north than in
summer 2003. Regions such as Northern France,
Germany, Belgium, and the Czech Republic,
which were strongly affected by the 2003 heatwave, were affected by an even more extreme
heatwave in July 2006, in terms of absolute mean
monthly values.
The statistical distribution of monthly July
temperature values in France (Fig. 3) shows that
July 2006 was by far the hottest month of July
for both minimum and maximum temperature.
The variability is higher for maximum than for
minimum temperature. In Switzerland, July
2006 was also by far the hottest month of July
ever measured. Both minimum and maximum
temperature distributions have a positive asymmetry coefficient.
4. Discussion
In July 2006, the heat was linked to persistent
anticyclonic situations favouring the advection
of dry air masses. A ridge extended at the
500 hPa level, from Spain to Denmark, during
most of the month (1st –5th July, 10th –27th July)
causing an Omega blocking situation (15th –27th
July), which forced the jet-stream to flow well
north of Scotland. On July 27th, the westerlies
resumed, bringing a frontal zone across
Western Europe.
In summer 2003, anticyclonic weather types also dominated over Central Europe, especially in
spring and summer (Black et al. 2004; Fink et al.
2004; Schaer et al. 2004). A large-scale anticyclonic forcing cannot alone explain extreme surface air temperature values as several factors such
as heat advection, radiation, cloud cover, moist
convection and other processes can vary between
different periods of the same anticyclonic weather
type. However the prevalence of anticyclonic
weather leads to a lack of soil moisture, which
generally favours high temperatures by enhancing
the surface sensible heat flux and substantially reducing the surface latent heat flux. This process
may be amplified by a positive feedback between
soil moisture and precipitation during heatwaves
(Seneviratne et al. 2006; Della-Marta et al. 2007b;
Fischer et al. 2007). In conjunction with these
drought conditions, a high level of radiation is
expected to increase maximum temperatures.
North-Eastern France, for example, experienced
record-breaking sunshine hours in June 2003,
was very sunny, although not record-breaking, in
July 2003, and had the largest record-breaking
number of sunshine hours in August 2003
(Rebetez et al. 2006). In July 2006, new recordbreaking sunshine durations were measured at all
stations located in North-Eastern France. For the
three months studied here, the anomalies are clearly larger for maximum than for minimum temperature. The discrepancy between minimum and
maximum temperature anomalies was even larger
in 2006 compared to both June and August 2003.
For the future, results of climate models suggest
that temperature variability should be increasing
for summer temperatures in a warmer climate
due to the impact of lower soil humidity (Schaer
et al. 2004; Seneviratne et al. 2006).
The statistical distribution of monthly July
temperatures (Fig. 2) shows that although only
the north-eastern part of the French territory
was most affected by the 2006 anomalies, July
2006 was by far the hottest month for both minimum and maximum temperature. The variability
is clearly higher for maximum temperature than
for minimum temperature, but the positive asymmetry coefficient of the distribution shows that
the variability of the warmer part of the series
is higher than that of the colder part for both
minimum and maximum temperature.
5. Conclusions
In July 2006, as in June and August 2003, mean
temperature anomalies reached more than 6 K
over large parts of Europe. Most anomalous
for all 3 months were maximum temperatures,
based on the affected areas, as well as on the
absolute values of the anomalies. Considering
the areas affected by the maximum temperature
anomalies, July 2006 was the most anomalously
hot summer month ever measured in Europe:
686,000 km2 were affected by a maximum temperature anomaly above 6 K and 2.21 million
km2 by an anomaly above 4 K. Considering the
monthly anomalies, June and August 2003 were
M. Rebetez et al.
more extreme, but based on monthly absolute
temperature values, our results show that for
the regions affected by the 2003 and 2006 heatwaves, more severe temperature conditions occurred in July 2006 in the north of Europe.
It is likely that the higher anomalies measured
in our three case studies for maximum temperature as compared to minimum temperature were
due to particularly low values in soil humidity,
hampering a cooling of the ground by evaporation. The extremely high values reached in 2006
after the exceptional summer 2003 should be understood as one more sign indicating that an increase in temperature variability could lead to
particularly hot summer temperatures, as maximum temperatures during heatwaves will tend to
be hotter than expected. In a warmer climate,
heatwaves will probably be more extreme than
expected, particularly during daytime, due to increasing summer temperature variability.
Acknowledgements
We are grateful to Meteo-France and MeteoSwiss for providing the climatological data, to Elena Rossel and Jacques
Ferrez for their technical help with the data, and to John
Innes for useful comments and suggestions. This work has
been supported in part by the Forest Investigation Program, a
joint project between the Swiss Federal Office of the
Environment (BAFU) and the Swiss Federal Institute for
Forest, Snow and Landscape Research (WSL).
References
Barnett J (2001) Security and Climate Change. Working
paper no. 7. Report No. 7, Tyndall Centre for Climate
Change Research
Bell JL, Sloan LC, Snyder MA (2004) Regional changes in
extreme climatic events: a future climate scenario. J
Climate 17: 81–87
Black E, Blackburn M, Harrison G, Hoskins B, Methven J
(2004) Factors contributing to the summer 2003 European
heatwave. Weather 59: 217–223
De US, Khole M, Dandekar MM (2004) Natural hazards
associated with meteorological extreme events. Nat
Hazards 31: 487–497
Della-Marta PM, Haylock MR, Luterbacher J, Wanner H
(2007a) Doubled length of western European summer
heat waves since 1880. J Geophys Res Atmosph 112:
D15103
Della-Marta PM, Luterbacher J, von Weissenfluh H, Xoplaki
E, Brunet M, Wanner H (2007b) Summer heat waves
over western Europe 1880–2003, their relationship to
large-scale forcings and predictability. Clim Dynam 29:
251–275
Easterling DR, Evans JL, Groisman PY, Karl TR, Kunkel
KE, Ambenje P (2000a) Observed variability and trends in
extreme climate events: a brief review. Bull Amer Meteor
Soc 81: 417–425
Easterling DR, Karl TR, Gallo KP, Robinson DA, Trenberth
KE, Dai A (2000b) Observed climate variability and
change of relevance to the biosphere. J Geophys ResAtmosph 105: 20101–20114
Easterling DR, Meehl GA, Parmesan C, Changnon SA,
Karl TR, Mearns LO (2000c) Climate extremes: observations, modeling, and impacts. Science 289:
2068–2074
Fink A, Brucher T, Kruger A, Leckebusch GC, Pinto JG,
Ulbrich U (2004) The 2003 European summer heatwaves
and drought – synoptic diagnosis and impacts. Weather
59: 2009–2216
Fischer EM, Seneviratne SI, Luthi D, Schaer C (2007)
Contribution of land-atmosphere coupling to recent
European summer heat waves. Geophys Res Lett 34:
L06707
Haines A, Patz JA (2004) Health effects of climate change.
J Am Med Assoc 291: 99–103
IPCC (2001) Climate change 2001. The Scientific Basis.
IPCC, Cambridge
IPCC (2002) Climate change 2001: Mitigation., Vol. 3.
Cambridge University Press, Cambridge, UK
IPCC (2007) Climate change 2007. The Scientific Basis,
IPCC, Cambridge
Kovats RS, Haines A (2005) Global climate change and
health: recent findings and future steps. Can Med Assoc J
172: 501–502
Kunkel KE, Pielke RA, Changnon SA (1999) Temporal
fluctuations in weather and climate extremes that cause
economic and human health impacts: a review. Bull Amer
Meteor Soc 80: 1077–1098
Luterbacher J, Dietrich D, Xoplaki E, Grosjean M, Wanner
H (2004) European seasonal and annual temperature
variability, trends, and extremes since 1500. Science
303: 1499–1503
McMichael A, Woodruff R (2004) Climate change and risk
to health. Br Med J 329: 1416–1417
McMichel AJ, Haines A, Slooff R, Kovats S (2004) Climate
change and human health. WHO, WMO, UNEP,
Washington
Meehl GA, Karl T, Easterling DR, Changnon S, Pielke R,
Changnon D, Evans J, Groisman PY, Knutson TR, Kunkel
KE, Mearns LO, Parmesan C, Pulwarty R, Root T, Sylves
RT, Whetton P, Zwiers F (2000) An introduction to trends
in extreme weather and climate events: observations,
socioeconomic impacts, terrestrial ecological impacts,
and model projections. Bull Amer Meteor Soc 81:
413–416
Menne B, Bertollini R, Kovats S, Lindgren E, Daniel M, Kriz
B, Ebi K, Hajat S, Edwars S, Haines A, Alberini A,
Chiabai A, van Lieshout M, Martens P, Jendritzky G,
Koppe C, Klein R, Fuessal M (2003) European climate
change health impact and adaptation assessment. Epidemiology 14: S116–S116
Murnane RJ (2004) Climate research and reinsurance. Bull
Amer Meteor Soc 85: 697–707
An analysis of the July 2006 heatwave extent in Europe compared to the record year of 2003
Parmesan C, Root TL, Willig MR (2000) Impacts of extreme
weather and climate on terrestrial biota. Bull Amer Meteor Soc 81: 443–450
Rebetez M (2004) Summer 2003 maximum and minimum
daily temperatures over a 3300 m altitudinal range in the
Alps. Climate Res 27: 45–50
Rebetez M, Mayer H, Dupont O, Schindler D, Gartner
K, Kropp J, Menzel A (2006) Heat and drought
2003 in Europe: a climate synthesis. Ann For Sc 63:
569–577
Schaer C, Vidale PL, Luthi D, Frei C, Haberli C, Liniger
MA, Appenzeller C (2004) The role of increasing temperature variability in European summer heatwaves. Nature 427: 332–336
Schnur R (2002) Climate science: the investment forecast.
Nature 415: 483–484
Sch€onwiese CD, Staeger T, Tromel S (2004) The hot
summer 2003 in Germany. Some preliminary results
of a statistical time series analysis. Meteorologische
Zeitschrift 13: 323–327
Seneviratne SI, Luthi D, Litschi M, Schaer C (2006) Landatmosphere coupling and climate change in Europe. Nature 443: 205–209
Smoyer KE (1998) A comparative analysis of heat waves
and associated mortality in St. Louis, Missouri – 1980 and
1995. Int J Biometeorol 42: 44–50
Sparks J, Changnon D, Starke J (2002) Changes in the
frequency of extreme warm-season surface dewpoints
in northeastern Illinois: implications for cooling-system
design and operation. J Appl Meteor 41: 890–898
Stephenson J (2004) Climate change and health. J Am Med
Assoc 291: 291–291
Stott PA, Stone DA, Allen MR (2004) Human contribution to
the European heatwave of 2003. Nature 432: 610–614
Trigo RM, Garcia-Herrera R, Diaz J, Trigo IF, Valente MA
(2005) How exceptional was the early August 2003
heatwave in France? Geophys Res Lett 32: L10701
Vescovi L, Rebetez M, Rong F (2005) Assessing public
health risk due to extremely high temperature events:
climate and social parameters. Climate Res 30: 71–78